ȧyių πνeúμаTI. We are told that the cautious old Arian, Leontius, when intruded into the see of Antioch, only raised his voice, when saying the Gloria, at the words 'world without end', and was indistinctly heard for all the earlier part of the formula. St Basil's use of the more archaic form at Caesarea gave occasion to suspicions of his orthodoxy and so indirectly to the composition of his best known work, the de Spiritu sancto. As St Basil demonstrated, both forms are in themselves perfectly orthodox and legitimate, and we must not assume that only Arians used the Sià... èv, though it is safe to assume that only Catholics used μerà...σúv. Therefore, if the text of the Constitutions employed a μerá form of doxology, we could hardly place the date of their composition before the very end of the fourth century, since they are certainly Antiochene, and it is quite unlikely that the Nicene doxology should have come into use at Antioch before the time of the emperor Theodosius-if indeed as soon as that. Now the MSS (at least in the eighth book) do not hold consistently to either form; five times at least -I take my data from Lagarde's text and apparatus-they agree on the formula δι' οὗ σοὶ . . . ἐν ἁγίῳ πνεύματι, 241. 8, 242. 11, 243. Ι, 244. 24, 247. 23; more than twice as often they agree on the alternative type, μεθ ̓ οὗ σοι . . . καὶ τῷ ἁγίῳ πνεύματι ; twice they approach the still more definite Western statement, 258. 6 τῷ πατρὶ καὶ τῷ υἱῷ καὶ τῷ ἁγίῳ πνεύματι, and 261. 20 σοὶ . . . καὶ τῷ σῷ παιδὶ . . . καὶ τῷ ἁγίῳ πνεύματι. In four cases the MSS differ among themselves; in these four cases t always gives the type δι' οὗ σοὶ . . . ἐν ἁγίῳ πνεύματι, in two of them, 274. 18, 275. 20, both x y have the μerà... κaí type, in the other two, 239. 1, 272. 16, y goes with t, while x gives the unique forms μεθ' οὗ καὶ δι ̓ οὗ . . . ἐν ἁγίῳ πνεύματι, δι ̓ οὗ σοὶ . . . σὺν ἁγίῳ πνεύματι. Lagarde consistently follows x: but I have no doubt at all that where t, or yt, give the form dià . . . ¿v their reading ought to be accepted. And more than that, the Latin version supplies good ground for suspecting that the μerà. . . κaí type, often as it occurs in all our Greek MSS, is due to revision; for in 276. 2 txy have μerà... Kaí, while the Latin (p. 494 l. 21) has nevertheless 'per quem . . . in sancto spiritu'. The only alternative to accepting the Latin evidence as decisive-the Latin MS, we know, is many centuries earlier than any of our Greek MSS-would be to suppose that the Latin translator wrote from a definitely Arian point of view, and refused to use the Nicene doxology. ... ... With w, which is by two centuries the oldest of Lagarde's MSS, the Latin has, in the brief space where both are extant, not less than five or six agreements against x y, with t sometimes on one side sometimes on the other: 496. 16 bibere with w wível, against Tive tx y; 497. 7 verba with tw τῶν λόγων, against τὸν λόγον x y ; 497. 7 suscipi with w προσλαμβάνεσθαι, against προσλαμβάνεσθε txy; 497. 13 usum with t w χρείαν, against χρείας xy; 497. 18 hoc autem with w τοῦτο δέ, against ἐκεῖνο δέ txy. To these we ought to add 495. 2, where w has ev yaλuoîs kai προσευχαῖς, the rest ἐν ψαλμοῖς καὶ ἀναγνώσεσιν (ἀναγνώσμασιν) καὶ Tроσevɣaîs, while the Latin, as Dr Spagnolo has now been able to decipher it, gives only 'psalms and prayer'. Of course both t and w have their own idiosyncrasies and blunders not shared by the Latin version; but speaking generally the Latin, as between the Greek witnesses, agrees with the older and better of them. We have now to ask what is the value of the Latin as against the whole of the extant Greek evidence, and we have already had reason, in connexion with the form of doxology, to suppose that the Latin, even when standing alone, may sometimes outweigh all the rest. Much the most important difference between the Latin version and Lagarde's edition is that in the version the Constitutions are immediately followed by the Canons. But Lagarde himself (p. 284) explains that, though absent from xy, the Canons had a place in w as well as in the edition of Turrianus; and probably the real reason why Lagarde included the Constitutions alone in his edition was that he had already published the text of the Apostolic Canons on pp. 20-35 of his Reliquiae, so that he is content to give (pp. 285-287) a collation of the differences of these two witnesses from his earlier text. Unfortunately the text of the Canons is very difficult, in some places impossible, to decipher in our MS, and I do not attempt to print the transcription in the JOURNAL, as it will find a more appropriate place in Ecclesiae Occidentalis Monumenta Iuris Antiquissima. I may, however, state here that, whereas Dionysius Exiguus only translated the first fifty canons (Eccl. Occ. Mon. Iur. Ant. i 1-32), the whole number are present in the Verona MS, and it is in the last canon, the list of Scriptural books, that the Latin version makes its most interesting contribution to the critical apparatus of the Canons. The text of this list will be found on pp. 511-514 of the present number of the JOURNAL, as the fourth of my series of 'Latin lists of the Canonical Books'. Another characteristic feature which distinguishes our version from Lagarde's edition is probably again more apparent than real. At 494. 25, 497. 3, 497. 14, the reader will find chapter-headings in the Latin (distinguished for his convenience by heavy type) to which there is nothing in the parallel text from Lagarde to correspond. But if he looks up Lagarde's preface (p. vi ad fin.) he will discover that that illustrious but self-willed scholar found chapter-headings in his MSS, but, in the conviction that they were not original, declined to put them in the text or even to record them in the apparatus; 'capita mea eadem sunt ac Turriani . . . capitum titulos a monachis profectos edere necessarium non videbatur.' A more modest procedure would have been triumphantly justified by the Latin; for the three chapter-headings it preserves coincide with the commencement of chapters 42 (42 43 44 are closely connected in subject matter, and form as it were only a single section), 45, and 46 in Lagarde, and I have little doubt that they are faithfully rendered from the Greek. I do not indeed see any prima facie reason why they should be attributed to 'monks' rather than to the compiler of the Constitutions himself. These more general considerations may be fitly followed by a briefer notice of individual features (a) in the Greek readings underlying the text, (b) in the Latinity of the translation. a. At p. 495 1. 6 I had ventured to restore 'tricesima' to the text; I could not make the indications of the MS reading square with quadragesima, and the 'ancient type' to which appeal is made, the mourning for Moses, lasted not forty but thirty days, hence I conjectured that the compiler wrote Tρiakoσтά, and that the Latin alone now bore witness to the original reading. On a second inspection of the MS Dr Spagnolo is able to make out nearly the whole of the word 'tricensima'. At p. 502 1. 24 the Latin has 'ego Clemens ego Iacob' for the Greek ἐγὼ Ἰάκωβος καὶ ἐγὼ Κλήμης: and we have to balance the respective probabilities that a Latin translator altered his model in order to do honour to Clement of Rome, and that Greek scribes altered their exemplars to do honour to James of Jerusalem. On the whole I think there is more to be said this time for the Greek texts. At p. 503 1. 2 the orders enumerated are in the Latin 'presbyters, deacons, and readers', in the Greek 'presbyters, deacons, subdeacons, and readers'. It would be interesting to suppose the Latin text right; but subdeacons are recognized elsewhere in the Constitutions, and probably the true explanation is either that it is a mere omission, by translator or scribe, through homoeoteleuton (diákovoι Kaì vπоdiákovoι, diaconi et subdiaconi), or that the Latin translator was not acquainted in his own region with the order of subdeacons, and simply omitted the name and thing. At Rome and Carthage subdeacons, hypodiaconi, were a part of the clerus as early as the middle of the third century; but none of the bishops who subscribed at the council of Arles in 314 brought any attendants other than presbyters or deacons or exorcists or readers, and the canons of Sardica enumerate only the orders of bishop, presbyter, deacon, and reader. If our translator wrote about A. D. 400, and in north-eastern Italy, he may perhaps have been equally unfamiliar with the subdiaconate. b. In the Latinity of the translation there is entire absence of any trace of influence of the Vulgate; but it is fair to add that there is hardly more evidence of Old Latin renderings. The translator must have worked at his task with the single preoccupation of representing the Greek original by Latin equivalents of his own; no other explanation will account for the strange 'spiritus paratus caro autem debilitate laborat' (497. 8) where the old biblical version 'spiritus promptus caro autem infirma' was left untouched by Jerome, just as 'the spirit indeed is willing but the flesh is weak' survives unaltered in the Revised Version of 1881. The representation of Aevîraɩ by ‘diaconi', 500. 16, excludes a date earlier than the end of the fourth century, but then the date of the Greek Constitutions themselves gives a similar terminus a quo; and in view of the tendency in writers like Ambrosiaster to identify the three orders in the Jewish hierarchy with the three orders of the Christian ministry, I should not like to say that 'diaconus' for Aevírns is necessarily later than 400. The converse use of 'levita' for deacon occurs already in pope Damasus and St Ambrose. But the most primitive sounding term employed by our translator is Elevatio' (503. 15) for the Ascension, Greek áváλns. Unless this is a pure coinage out of his own head-on the same lines as his 'debilitate laborat'-it suggests a very archaic terminology for the Christian festivals. It does not seem to be paralleled at all elsewhere: 'ascension' (in one or other form of the word) is the universal Latin term. 'Pontifex' for åpxiepeús (p. 504 ll. 24, 27) may also suggest an early date; after the end of the fourth century 'summus sacerdos' or 'princeps sacerdotum became the fashionable ways to render 'high priest' as opposed to 'priest'. Of the manuscript in which the fragment is contained I spoke in the October number of the JOURNAL (pp. 19-28 of the present volume), and I need only add that the exemplar of this part of the MS must have been written in lines of about twenty-one or twenty-two letters-rather shorter, that is, than in our MS, which averages some twenty-five or twenty-six letters-as is indicated by the two omissions on p. 493 ll. 6, 18. The pages here transcribed and printed are at places extraordinarily difficult to read, and the italic type represents the best reconstruction I could make out of the data supplied by Dr Spagnolo, to whom scholars are again indebted for the time and zeal which he has devoted to the work of decipherment. Angular brackets() denote that in those passages the MS is not only worn but worn or torn away, so that the supplements are necessarily conjectural. C. H. TURNER. LATIN LISTS OF THE CANONICAL BOOKS. IV. AN EARLY VERSION OF THE EIGHTY-FIFTH FROM MS VERON. LI foll. 1556, 156 a. Ar two points in the current volume of the JOURNAL, pp. 19 ff of the October number and p. 510 of the present number, a description has been given of the Verona MS from which the following list of the Canonical Books is derived. As I have already said on p. 508, the Latin fragment of the end of the eighth book of the Apostolic Constitutions is followed without break by the Apostolic Canons; and since Dionysius Exiguus, for what reason I know not, only translated the first fifty canons, our MS is the only early authority for the remaining thirtyfive-and therewith for the last of all, containing the Biblical list. The terminus a quo for the date of the version contained in our MS is the date of the Apostolic Constitutions and Canons themselves, which may be put at about A.D. 400 or a little earlier; the terminus ad quem is the date of our Latin MS, and that can hardly be after A. D. 600. In my own opinion the version is probably nearer the earlier than the later of the two termini; but in any case both the version and the MS in which it is contained are probably older by some centuries than the oldest Greek MSS hitherto employed. The differences between the Latin and the printed Greek texts are, as will be seen on comparison of the two columns of p. 513, sufficiently serious; they would have been considerably more serious if I had (as in the fragment of the Constitutions, pp. 492 sqq. supra) chosen the text of Lagarde for the Greek column. But the text of Turrianus is here so manifestly superior to that of Lagarde that the only reasonable course appeared to be to give it on this occasion the preference. Lagarde omits the book of Judith, the book of Job, the Psalms; reduces the books of Maccabees from three to one; conversely makes the Wisdom of Sirach into Wisdoms; and adopts a form of doxology with our instead of iv. In these and other points Turrianus agrees with the Latin, and he would be a bold critic who ventured to maintain in any one of them the superiority of the reading of Lagarde.' Tobit is omitted in all the texts, Turrianus, Lagarde, and Latin alike. More interest attaches to the divergences between the text of Turrianus 1 On the doxology see above, p. 506. |